@inproceedings{7173, abstract = {Reinforcement learning has achieved state-of-the-art performance in MAV control, waypoint flight, and obstacle avoidance. However, existing RL approaches often assume fixed objectives and constraints, with flight behavior largely limited by vehicle dynamics and orientation considered only when required for locomotion. Classical planning and model predictive control handle such constraints explicitly, but require optimization or replanning. This motivates methods that combine learned local control with explicit constraint handling. We combine reinforcement learning with control barrier functions to improve constraint-aware execution. We propose parameterized waypoints that encode orientation, velocity, and corridor constraints. Simulation and real-world experiments show that a single policy can execute different constraint-parameterized navigation scenarios, revealing scenario-dependent trade-offs between traversal time, tracking accuracy, and constraint satisfaction.}, author = {Kirsch, André and Rexilius, Jan}, keywords = {MAV navigation, Reinforcement learning, Constrained navigation, Control barrier functions}, location = {Barcelona}, title = {{Reinforcement Learning-based MAV Navigation With Parameterized Waypoints: Incorporating Orientation, Speed Limits, and Corridor Constraints}}, year = {2026}, } @inproceedings{7174, abstract = {This paper presents a hybrid control architecture for dynamic robotic picking tasks. The framework combines a Deep Reinforcement Learning policy for high-level interception with a dedicated Inverse Kinematics controller for precise terminal grasping, while mitigating precision limitations of monolithic learning-based approaches. The framework utilizes a Proximal Policy Optimization agent to approach moving targets, seamlessly transitioning to an Inverse Kinematics solver that reduces terminal orientational and positional errors while minimizing cumulative control effort. To facilitate deployment on physical hardware, a robust sim-to-real pipeline incorporating system identification, domain randomization, and latency injection is employed. Experimental results on a physical Franka Emika Panda manipulator validate this hybrid architecture. The system achieves an 80% success rate in pick-and-place tasks, compared to 60.8% for unadapted baselines, with no safety-critical violations such as joint limit breaches or collisions observed during testing.}, author = {Mayer, Patrick Thomas and Rexilius, Jan}, keywords = {Reinforcement Learning, Hybrid Control, Sim-to-Real, Robotic Manipulation, Inverse Kinematics.}, location = {Sapporo}, title = {{A Hybrid Control Framework Using Reinforcement Learning for Dynamic Robotic Manipulation and Sim-to-Real Transfer}}, year = {2026}, } @inproceedings{7175, abstract = {Deep reinforcement learning (RL) policies for robotic control typically overfit to a single hardware configuration. Any change to the kinematic chain breaks the learned mapping and requires complete retraining. This paper investigates whether combining self-attention mechanisms with explicit spatial observations can improve a policy’s adaptability to varying kinematic topologies. We train a single RL agent to control a robotic manipulator across different degrees of freedom (DOF), ranging from a restricted 4-DOF mode to a fully redundant 7-DOF configuration. Instead of fixed-length state vectors, the method processes the active joints as a variable-length sequence in an attention buffer, enriching each joint’s representation with its relative spatial routing and geometric Jacobian influence. This structure allows the agent to dynamically evaluate the physical utility of its available actuators. The proposed Architecture reaches a success rate of 82.1% across trained topologies and 66.6% on unseen configurations, outperforming MLP and generic attention baselines while remaining nearly collisionfree. Finally, we demonstrate successful sim-to-real transfer by deploying the simulation-trained agent on a physical Franka Emika Panda manipulator.}, author = {Mayer, Patrick Thomas and Rexilius, Jan}, keywords = {Reinforcement Learning, Robotic Manipulation, Morphology Generalization, Sim-to-Real Transfer}, location = {Barcelona}, title = {{Adapting to Variable Kinematic Configurations: A Causal-Kinematic Attention Approach for Robotic Control}}, year = {2026}, } @inproceedings{7068, abstract = {Adopting digital twin technologies in small andmedium-sized enterprises (SMEs) is often hinderedby heterogeneous, poorly documented productiondata. Current semantic type detection approachesrequire massive labeled datasets making themimpractical for resource-constrained SMEs. Wepropose a zero-shot hybrid framework combiningpattern-based classification with selective largelanguage model (LLM) reasoning formanufacturing-specific data types. The two-stagearchitecture uses rule-based patterns for high-confidence cases, forwarding ambiguous columnsto a multi-step LLM reasoner. Evaluation on fourmanufacturing datasets shows the hybrid approachachieves weighted F1 within 7 24 percentagepoints of pure LLM classification performancewhile reducing LLM invocations by 39% onaverage. Processing time decreased by up to 2.8×.Our framework addresses a critical gap: automated,computationally efficient data type recognition formanufacturing SMEs without requiring trainingdata, contributing to automated simulation anddigital twin construction.}, author = {Döring, Lina and Trojahn, Sebastian and Reusch, Pascal}, booktitle = {19th International Doctoral Students Workshop on Logistics, Supply Chain and Production Management}, editor = {Behrendt, Fabian and Zadek, Hartmut and Janmontree, Jettarat and Trojahn, Sebastian and Lang, Sebastian}, keywords = {Semantic Data Type Detection, DigitalTwin, Manufacturing SMEs, Zero-shot Learning, Hybrid AI, LLM}, location = {Magdeburg}, publisher = {Otto von Guericke University Library, Magdeburg, Germany}, title = {{Smart data adapter: A hybrid pattern-LLM approach}}, doi = {10.25673/123544}, year = {2026}, } @inproceedings{6235, author = {Sangel, Marius and Bensch, Emilia and Brandt-Pook, Hans and Röllke, Timo and Markworth, Cedric}, booktitle = {INFORMATIK 2025}, keywords = {Machine Learning, Computer Vision, Instance Segmentation, CNNs, YOLACT, Data Augmentation, Waste Classification, Trash Detection, Biowaste Analysis}, location = {Potsdam}, number = {366}, pages = {1363--1371}, title = {{Automatisierte Erkennung von Störstoffen in Bioabfall mit maschinellem Lernen: Ansätze und Ergebnisse aus dem Projekt TRACES}}, doi = {10.18420/INF2025_121}, year = {2025}, } @inproceedings{5905, author = {Jaster, Bjarne and Kohlhase, Martin}, booktitle = {2025 IEEE Symposium on Trustworthy, Explainable and Responsible Computational Intelligence (CITREx Companion)}, keywords = {Active Learning, Trustworthiness, Evaluation-methods}, location = {Trondheim, Norway}, pages = {1--5}, publisher = {IEEE}, title = {{Trust Issues in Active Learning and Their Impact on Real-World Applications}}, doi = {10.1109/CITRExCompanion65208.2025.10981492}, year = {2025}, } @inproceedings{6045, author = {Jaster, Bjarne and Kohlhase, Martin}, booktitle = {2025 IEEE Symposium on Trustworthy, Explainable and Responsible Computational Intelligence (CITREx Companion)}, keywords = {Active Learning, Trustworthiness, Evaluation-methods}, location = {Trondheim, Norway}, pages = {1--5}, publisher = {IEEE}, title = {{Trust Issues in Active Learning and Their Impact on Real-World Applications}}, doi = {10.57720/6045}, year = {2025}, } @inproceedings{6049, author = {Schöne, Marvin and Jaster, Bjarne and Bültemeier, Julian and Kösters, Justus and Holst, Christoph-Alexander and Kohlhase, Martin}, booktitle = {2025 IEEE Symposium on Trustworthy, Explainable and Responsible Computational Intelligence (CITREx)}, keywords = {Interpretability, Classification, Pool-based Active Learning, Decision Trees}, location = {Trondheim, Norway}, pages = {1--9}, publisher = {IEEE}, title = {{Pool-based Active Learning with Decision Trees: Incorporate the Tree Structure to Explore and Exploit}}, doi = {10.57720/6049}, year = {2025}, } @inproceedings{5904, author = {Schöne, Marvin and Jaster, Bjarne and Bültemeier, Julian and Kösters, Justus and Holst, Christoph-Alexander and Kohlhase, Martin}, booktitle = {2025 IEEE Symposium on Trustworthy, Explainable and Responsible Computational Intelligence (CITREx)}, keywords = {Interpretability, Classification, Pool-based Active Learning, Decision Trees}, location = {Trondheim, Norway}, pages = {1--9}, publisher = {IEEE}, title = {{Pool-based Active Learning with Decision Trees: Incorporate the Tree Structure to Explore and Exploit}}, doi = {10.1109/CITREx64975.2025.10974940}, year = {2025}, } @phdthesis{4392, author = {Grumbach, Felix}, keywords = {Produktionsplanung und -steuerung, Operations Research, Machine Learning, Reinforcement Learning}, publisher = {Universitäts- und Landesbibliothek Sachsen-Anhalt}, title = {{Feldsynchrone Ablaufplanung dynamischer Fertigungsprozesse mit Techniken des maschinellen Lernens [kumulative Dissertation]}}, doi = {10.25673/115290}, year = {2024}, } @article{4881, author = {Wiegraebe, Frauke and Schönbeck, Marleen and Wunderlich, Paul and Nauerth, Annette and Dörksen, Helene}, issn = {1430-9653}, journal = {Pflege und Gesellschaft}, keywords = {Digital care support tool, caring workforce, interdisciplinarity, AI application, machine learning, user orientation}, pages = {271--285}, publisher = {BeltzJuventa}, title = {{KI-basiertes Unterstützungstool für pflegende Erwerbstätige}}, doi = {10.3262/P&G2403271}, volume = {3}, year = {2024}, } @article{2767, author = {Richter, Niclas and Tuvshinbayar, Khorolsuren and Ehrmann, Guido and Ehrmann, Andrea}, issn = {2079-9292}, journal = {Electronics}, keywords = {smart textiles, smart clothes, e-textiles, biosignal, vital signal, pulse sensor, microcontroller, washability, display}, number = {7}, publisher = {MDPI AG}, title = {{Usability of Inexpensive Optical Pulse Sensors for Textile Integration and Heartbeat Detection Code Development}}, doi = {10.3390/electronics12071521}, volume = {12}, year = {2023}, } @techreport{3729, author = {Kösters, Justus and Schöne, Marvin and Kohlhase, Martin}, keywords = {tabular scarce data, industrial design, supervised machine learning models}, title = {{Benchmarking of Machine Learning Models for Tabular Scarce Data}}, year = {2023}, } @article{4073, author = {Armutat, Sascha and Mauritz, Nina and Wattenberg, Malte and Bormann, Frank}, issn = {1618-8543}, journal = {bwp@ Berufs- und Wirtschaftspädagogik – online}, keywords = {Learning-Management-System, Web-Based-Training, Teilnehmerorientierung, Technologieakzeptanz}, number = {45}, pages = {1--22}, title = {{Digitalisiertes Lernen in der be­trieblichen Weiterbildung – zielgruppenorientierte Akzeptanz-Anforderungen an virtuelle Pro­duktschulungen am Beispiel der Weidmüller Interface GmbH & Co. KG}}, year = {2023}, } @techreport{2692, author = {Strecker, Mia Jasmin and Freese, Christiane and Nagel, Lisa and Stirner, Alexander and Freese, Isa and Oldak, Anna and Hejna, Urszula and Pieper, Melanie and Hainke, Carolin and Lätzsch, Rebecca and Wattenberg, Ivonne and Falk-Dulisch, Miriam and Liebau, Laura and Eickelmann, Anne-Kathrin and Hornberg, Claudia and Pfeiffer, Thies and Seeling, Stefanie and Kamin, Anna-Maria and Makowsky, Katja and Nauerth, Annette}, keywords = {Blended Learning, Pflege, Medizin, Fachdidaktik, Digitale Lehre, fallbasiertes Lernen}, pages = {228}, title = {{Didaktisches Fachkonzept zur digitalen und virtuell unterstützten Fallarbeit in den Gesundheitsberufen}}, doi = {10.57720/2692}, year = {2023}, } @inproceedings{4206, author = {Sanaullah, Sanaullah and Amanullah, Amanullah and Roy, Kaushik and Lee, Jeong-A and Chul-Jun, Son and Jungeblut, Thorsten}, keywords = {hybrid SC-NN, SNN, CNN, Machine Learning}, location = {Paris France}, title = {{A Hybrid Spiking-Convolutional Neural Network Approach for Advancing High-Quality Image Inpainting}}, doi = {10.5281/zenodo.10458019}, year = {2023}, } @techreport{2763, author = {Fries, Sophia and Pfeifer, Lydia Sophie and Schlautmann, Katharina and Freese, Christiane and Nauerth, Annette and Raschper, Patrizia}, keywords = {Pflegeausbildung, SCRUM, Lernortkooperation, Lernaufgaben, Virtuelle Realität, Fortbildung, Nursing education, SCRUM, learning site cooperation, learning tasks, virtual reality, advanced training, divifag, virdipaworkingpaper}, publisher = {Universität Bielefeld}, title = {{Entwicklung und Erprobung digital gestützter Lernaufgaben mit VR-Szenarien. Das Fortbildungskonzept des Projekts ViRDiPA}}, doi = {10.4119/UNIBI/2978152}, year = {2023}, } @inproceedings{4394, author = {Müller, Anna and Grumbach, Felix}, booktitle = {16th International Doctoral Students Workshop on Logistics, Supply Chain and Production Management}, editor = {Glistau, Elke and Trojahn, Sebastian}, keywords = {Machine Learning, Production scheduling, time prediction}, location = {Magdeburg}, publisher = {Otto von Guericke University Library, Magdeburg, Germany}, title = {{Predicting processing times in high mix low volume job shops}}, doi = {10.25673/103491}, year = {2023}, } @inproceedings{4207, author = {Sanaullah, Sanaullah and Jungeblut, Thorsten}, keywords = {Brain Analysis, Machine Learning, Simulator, Runtime Simulator}, location = {New York USA}, title = {{Analysis of MR Images for Early and Accurate Detection of Brain Tumor using Resource Efficient Simulator Brain Analysis}}, doi = {10.5281/zenodo.10457930}, year = {2023}, } @techreport{3731, author = {Kösters, Justus and Schöne, Marvin}, keywords = {Scarce Data, Active Learning, GUIDEAlgorithmus, Modellunsicherheiten}, title = {{Active Learning mit dem GUIDE-Entscheidungsbaum}}, year = {2023}, }